🌐 2D Speaker Embedding Clusters: von Mises-Fisher Spread, Few-Shot Centroid Error and Bayes Overlap
Speaker embeddings as von Mises clouds on the unit circle: concentration, few-shot centroid arrows, nearest-centroid classification and the Bayes overlap of neighbouring clusters, with Bessel-function closed forms checked by integration and Monte Carlo (synthetic embeddings).
⚙ Under the hood
Speaker embeddings as von Mises clouds on the unit circle: concentration, few-shot centroid arrows, nearest-centroid classification and the Bayes overlap of neighbouring clusters, with Bessel-function closed forms checked by integration and Monte Carlo (synthetic embeddings).
Directional Statisticsvon Mises-FisherSpeaker EmbeddingCentroid EstimationBayes ErrorMean Resultant LengthUnit SphereClassification
2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install